Skip to content

Commit 981fbe3

Browse files
rflamaryrflamary
authored
[WIP] Build donc in GH Action and report warnings + move contributing and code of conduct in documentation (#441)
* use action for doc with wraning visible * remove space * remove space again * test pre commands * install pot properly * install compiler... * try composite action * remoe warning in sliced exmaple * pep8 * move contributing and code of conduct * cleanup * underline too short * update quickstart * replace version selector by static list to avoid jsQuery bug
1 parent 897026e commit 981fbe3

File tree

10 files changed

+134
-90
lines changed

10 files changed

+134
-90
lines changed

‎.github/CONTRIBUTING.md

-1
Original file line numberDiff line numberDiff line change
@@ -1,4 +1,3 @@
1-
21
Contributing to POT
32
===================
43

‎.github/workflows/build_doc.yml

+44
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,44 @@
1+
name: Build doc
2+
3+
on:
4+
workflow_dispatch:
5+
pull_request:
6+
push:
7+
branches:
8+
- 'master'
9+
10+
jobs:
11+
build:
12+
13+
runs-on: ubuntu-latest
14+
15+
steps:
16+
- uses: actions/checkout@v1
17+
# Standard drop-in approach that should work for most people.
18+
19+
- name: Set up Python 3.8
20+
uses: actions/setup-python@v1
21+
with:
22+
python-version: 3.8
23+
24+
- name: Get Python running
25+
run: |
26+
python -m pip install --user --upgrade --progress-bar off pip
27+
python -m pip install --user --upgrade --progress-bar off -r requirements.txt
28+
python -m pip install --user --upgrade --progress-bar off -r docs/requirements.txt
29+
python -m pip install --user --upgrade --progress-bar off ipython "https://api.github.com/repos/sphinx-gallery/sphinx-gallery/zipball/master" memory_profiler
30+
python -m pip install --user -e .
31+
# Look at what we have and fail early if there is some library conflict
32+
- name: Check installation
33+
run: |
34+
which python
35+
python -c "import ot"
36+
# Build docs
37+
- name: Generate HTML docs
38+
uses: rickstaa/sphinx-action@master
39+
with:
40+
docs-folder: "docs/"
41+
- uses: actions/upload-artifact@v1
42+
with:
43+
name: Documentation
44+
path: docs/build/html/

‎README.md

+1-1
Original file line numberDiff line numberDiff line change
@@ -192,7 +192,7 @@ POT has benefited from the financing or manpower from the following partners:
192192

193193
## Contributions and code of conduct
194194

195-
Every contribution is welcome and should respect the [contribution guidelines](.github/CONTRIBUTING.md). Each member of the project is expected to follow the [code of conduct](.github/CODE_OF_CONDUCT.md).
195+
Every contribution is welcome and should respect the [contribution guidelines](https://pythonot.github.io/master/contributing.html). Each member of the project is expected to follow the [code of conduct](https://pythonot.github.io/master/code_of_conduct.html).
196196

197197
## Support
198198

‎docs/source/_templates/versions.html

+36-33
Original file line numberDiff line numberDiff line change
@@ -1,47 +1,50 @@
11
<div class="rst-versions" data-toggle="rst-versions" role="note"
22
aria-label="versions">
3-
<!-- add shift_up to the class for force viewing -->
4-
<span class="rst-current-version" data-toggle="rst-current-version">
3+
<!-- add shift_up to the class for force viewing ,
4+
data-toggle="rst-current-version" -->
5+
<span class="rst-current-version" style="margin-bottom:1mm;">
56
<span class="fa fa-book"> Python Optimal Transport</span>
6-
versions
7-
<span class="fa fa-caret-down"></span>
7+
<hr style="margin-bottom:1.5mm;margin-top:5mm;">
8+
<!-- versions
9+
<span class="fa fa-caret-down"></span>-->
10+
<span class="rst-current-version" style="display: inline-block;padding:
11+
0px;color:#fcfcfcab;float:left;font-size: 100%;">
12+
Versions:
13+
<a href="https://pythonot.github.io/"
14+
style="padding: 3px;color:#fcfcfc;font-size: 100%;">Release</a>
15+
<a href="https://pythonot.github.io/master"
16+
style="padding: 3px;color:#fcfcfc;font-size: 100%;">Development</a>
17+
<a href="https://github.com/PythonOT/POT"
18+
style="padding: 3px;color:#fcfcfc;font-size: 100%;">Code</a>
19+
20+
</span>
21+
22+
823
</span>
9-
<div class="rst-other-versions"><!-- Inserted RTD Footer -->
24+
25+
<!--
26+
<div class="rst-other-versions">
27+
28+
1029
1130
<div class="injected">
1231
13-
14-
15-
<dl>
16-
<dt>Versions</dt>
17-
18-
<dd><a href="https://pythonot.github.io/">Release</a></dd>
19-
20-
<dd><a href="https://pythonot.github.io/master">Development</a></dd>
21-
22-
23-
24-
</dl>
25-
32+
33+
<dl>
34+
<dt>Versions</dt>
2635
36+
<dd><a href="https://pythonot.github.io/">Release</a></dd>
2737
28-
29-
<dl>
30-
<dt>On GitHub</dt>
31-
<dd>
32-
<a href="https://github.com/PythonOT/POT">Code on Github</a>
33-
</dd>
34-
35-
</dl>
36-
38+
<dd><a href="https://pythonot.github.io/master">Development</a></dd>
39+
3740
38-
39-
4041
41-
<hr>
42-
42+
<dt><a href="https://github.com/PythonOT/POT">Code on Github</a></dt>
4343
44+
45+
</dl>
46+
<hr>
4447
45-
</div>
46-
</div>
48+
</div>
49+
</div>-->
4750
</div>

‎docs/source/all.rst

+1-1
Original file line numberDiff line numberDiff line change
@@ -37,7 +37,7 @@ API and modules
3737

3838

3939
Main :py:mod:`ot` functions
40-
--------------
40+
---------------------------
4141

4242
.. automodule:: ot
4343
:members:

‎docs/source/.github/CODE_OF_CONDUCT.rst ‎docs/source/code_of_conduct.rst

+2-2
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
1-
Code of Conduct
1+
Code of conduct
22
===============
33

4-
.. include:: ../../../.github/CODE_OF_CONDUCT.md
4+
.. include:: ../../.github/CODE_OF_CONDUCT.md
55
:parser: myst_parser.sphinx_
66
:start-line: 2

‎docs/source/.github/CONTRIBUTING.rst ‎docs/source/contributing.rst

+2-2
Original file line numberDiff line numberDiff line change
@@ -1,6 +1,6 @@
11
Contributing to POT
22
===================
33

4-
.. include:: ../../../.github/CONTRIBUTING.md
4+
.. include:: ../../.github/CONTRIBUTING.md
55
:parser: myst_parser.sphinx_
6-
:start-line: 3
6+
:start-line: 2

‎docs/source/index.rst

+2-2
Original file line numberDiff line numberDiff line change
@@ -21,9 +21,9 @@ Contents
2121
all
2222
auto_examples/index
2323
releases
24-
.github/CONTRIBUTING
2524
contributors
26-
.github/CODE_OF_CONDUCT
25+
contributing
26+
code_of_conduct
2727

2828

2929
.. include:: ../../README.md

‎docs/source/quickstart.rst

+45-47
Original file line numberDiff line numberDiff line change
@@ -127,14 +127,6 @@ been used to solve both graph Laplacian regularization OT and Gromov
127127
Wasserstein [30]_.
128128

129129

130-
.. note::
131-
132-
POT is originally designed to solve OT problems with Numpy interface and
133-
is not yet compatible with Pytorch API. We are currently working on a torch
134-
submodule that will provide OT solvers and losses for the most common deep
135-
learning configurations.
136-
137-
138130
When not to use POT
139131
"""""""""""""""""""
140132

@@ -692,42 +684,8 @@ A list of the provided implementation is given in the following note.
692684
:heading-level: "
693685

694686

695-
Other applications
696-
------------------
697-
698-
We discuss in the following several OT related problems and tools that has been
699-
proposed in the OT and machine learning community.
700-
701-
Wasserstein Discriminant Analysis
702-
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
703-
704-
Wasserstein Discriminant Analysis [11]_ is a generalization of `Fisher Linear Discriminant
705-
Analysis <https://en.wikipedia.org/wiki/Linear_discriminant_analysis>`__ that
706-
allows discrimination between classes that are not linearly separable. It
707-
consists in finding a linear projector optimizing the following criterion
708-
709-
.. math::
710-
P = \text{arg}\min_P \frac{\sum_i OT_e(\mu_i\#P,\mu_i\#P)}{\sum_{i,j\neq i}
711-
OT_e(\mu_i\#P,\mu_j\#P)}
712-
713-
where :math:`\#` is the push-forward operator, :math:`OT_e` is the entropic OT
714-
loss and :math:`\mu_i` is the
715-
distribution of samples from class :math:`i`. :math:`P` is also constrained to
716-
be in the Stiefel manifold. WDA can be solved in POT using function
717-
:any:`ot.dr.wda`. It requires to have installed :code:`pymanopt` and
718-
:code:`autograd` for manifold optimization and automatic differentiation
719-
respectively. Note that we also provide the Fisher discriminant estimator in
720-
:any:`ot.dr.fda` for easy comparison.
721-
722-
.. warning::
723-
724-
Note that due to the hard dependency on :code:`pymanopt` and
725-
:code:`autograd`, :any:`ot.dr` is not imported by default. If you want to
726-
use it you have to specifically import it with :code:`import ot.dr` .
727-
728-
.. minigallery:: ot.dr.wda
729-
:add-heading: Examples of the use of WDA
730-
:heading-level: "
687+
Unbalanced and partial OT
688+
-------------------------
731689

732690

733691

@@ -845,10 +803,11 @@ regularization of the problem.
845803
:heading-level: "
846804

847805

806+
Gromov Wasserstein and extensions
807+
---------------------------------
848808

849-
850-
Gromov-Wasserstein
851-
^^^^^^^^^^^^^^^^^^
809+
Gromov Wasserstein(GW)
810+
^^^^^^^^^^^^^^^^^^^^^^
852811

853812
Gromov Wasserstein (GW) is a generalization of OT to distributions that do not lie in
854813
the same space [13]_. In this case one cannot compute distance between samples
@@ -877,6 +836,8 @@ There also exists an entropic regularized variant of GW that has been proposed i
877836
:add-heading: Examples of computation of GW, regularized G and FGW
878837
:heading-level: "
879838

839+
Gromov Wasserstein barycenters
840+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
880841

881842
Note that similarly to Wasserstein distance GW allows for the definition of GW
882843
barycenters that can be expressed as
@@ -905,6 +866,43 @@ The implementations of FGW and FGW barycenter is provided in functions
905866
:heading-level: "
906867

907868

869+
Other applications
870+
------------------
871+
872+
We discuss in the following several OT related problems and tools that has been
873+
proposed in the OT and machine learning community.
874+
875+
Wasserstein Discriminant Analysis
876+
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
877+
878+
Wasserstein Discriminant Analysis [11]_ is a generalization of `Fisher Linear Discriminant
879+
Analysis <https://en.wikipedia.org/wiki/Linear_discriminant_analysis>`__ that
880+
allows discrimination between classes that are not linearly separable. It
881+
consists in finding a linear projector optimizing the following criterion
882+
883+
.. math::
884+
P = \text{arg}\min_P \frac{\sum_i OT_e(\mu_i\#P,\mu_i\#P)}{\sum_{i,j\neq i}
885+
OT_e(\mu_i\#P,\mu_j\#P)}
886+
887+
where :math:`\#` is the push-forward operator, :math:`OT_e` is the entropic OT
888+
loss and :math:`\mu_i` is the
889+
distribution of samples from class :math:`i`. :math:`P` is also constrained to
890+
be in the Stiefel manifold. WDA can be solved in POT using function
891+
:any:`ot.dr.wda`. It requires to have installed :code:`pymanopt` and
892+
:code:`autograd` for manifold optimization and automatic differentiation
893+
respectively. Note that we also provide the Fisher discriminant estimator in
894+
:any:`ot.dr.fda` for easy comparison.
895+
896+
.. warning::
897+
898+
Note that due to the hard dependency on :code:`pymanopt` and
899+
:code:`autograd`, :any:`ot.dr` is not imported by default. If you want to
900+
use it you have to specifically import it with :code:`import ot.dr` .
901+
902+
.. minigallery:: ot.dr.wda
903+
:add-heading: Examples of the use of WDA
904+
:heading-level: "
905+
908906

909907
Solving OT with Multiple backends on CPU/GPU
910908
--------------------------------------------

‎examples/backends/plot_ssw_unif_torch.py

+1-1
Original file line numberDiff line numberDiff line change
@@ -50,7 +50,7 @@ def plot_sphere(ax):
5050
r = np.linspace(1.0, 1.0, 50)
5151
X, Y = np.meshgrid(xlist, ylist)
5252

53-
Z = np.sqrt(r**2 - X**2 - Y**2)
53+
Z = np.sqrt(np.maximum(r**2 - X**2 - Y**2, 0))
5454

5555
ax.plot_wireframe(X, Y, Z, color="gray", alpha=.3)
5656
ax.plot_wireframe(X, Y, -Z, color="gray", alpha=.3) # Now plot the bottom half

0 commit comments

Comments
 (0)